Autoencoding topology

Eric O. Korman

Autoencoding topology

Machine Learning 2018-03-02 v1 Machine Learning

Authors: Eric O. Korman

Abstract

The problem of learning a manifold structure on a dataset is framed in terms of a generative model, to which we use ideas behind autoencoders (namely adversarial/Wasserstein autoencoders) to fit deep neural networks. From a machine learning perspective, the resulting structure, an atlas of a manifold, may be viewed as a combination of dimensionality reduction and "fuzzy" clustering.

Keywords

manifold learning deep learning multi-task learning

Cite

@article{arxiv.1803.00156,
  title  = {Autoencoding topology},
  author = {Eric O. Korman},
  journal= {arXiv preprint arXiv:1803.00156},
  year   = {2018}
}

Comments

10 pages

Related papers

View all related →